Update app.py

app.py

@@ -17,10 +17,12 @@ st.set_page_config(page_title="Advanced Political Speech Analysis", page_icon="
 
 # Advanced NLP Libraries
 from transformers import (
-    AutoTokenizer, 
-    AutoModelForSequenceClassification, 
+    AutoTokenizer,
+    AutoModelForSequenceClassification,
     pipeline,
-    AutoModelForTokenClassification
+    AutoModelForTokenClassification,
+    RobertaTokenizer,
+    RobertaForSequenceClassification
 )
 import nltk
 from nltk.corpus import stopwords
@@ -58,12 +60,17 @@ RHETORICAL_DEVICES = {
 
 class SpeechAnalyzer:
     def __init__(self):
-        # Load models
-        self.moralbert_tokenizer = AutoTokenizer.from_pretrained("minyoungchang/moralbert")
-        self.moralbert_model = AutoModelForSequenceClassification.from_pretrained("minyoungchang/moralbert")
-        self.sentiment_pipeline = pipeline("sentiment-analysis")
+    # Load MoralFoundations model
+        self.moral_model_path = "MMADS/MoralFoundationsClassifier"
+        self.moral_tokenizer = RobertaTokenizer.from_pretrained(self.moral_model_path)
+        self.moral_model = RobertaForSequenceClassification.from_pretrained(self.moral_model_path)
+        
+        # Load label names
+        with open(f"{self.moral_model_path}/label_names.json", 'r') as f:
+            self.label_names = json.load(f)
         
-        # Named Entity Recognition
+        # Other pipelines remain the same
+        self.sentiment_pipeline = pipeline("sentiment-analysis")
         self.ner_tokenizer = AutoTokenizer.from_pretrained("dslim/bert-base-NER")
         self.ner_model = AutoModelForTokenClassification.from_pretrained("dslim/bert-base-NER")
         self.ner_pipeline = pipeline("ner", model=self.ner_model, tokenizer=self.ner_tokenizer)
@@ -90,29 +97,32 @@ class SpeechAnalyzer:
         return segments
 
     def analyze_moral_foundations(self, text):
-        """Analyze moral foundations in text"""
+    """Analyze moral foundations using the RoBERTa-based classifier"""
         segments = self.split_text(text)
         
         foundation_scores = {
             'care': [], 'fairness': [], 'loyalty': [],
             'authority': [], 'sanctity': []
         }
-
+        
         for segment in segments:
-            inputs = self.moralbert_tokenizer(segment, return_tensors="pt", truncation=True, max_length=512)
+            inputs = self.moral_tokenizer(segment, return_tensors="pt", truncation=True, max_length=512)
             
             with torch.no_grad():
-                outputs = self.moralbert_model(**inputs)
+                outputs = self.moral_model(**inputs)
             
             probabilities = torch.softmax(outputs.logits, dim=1)
             
-            for foundation in foundation_scores.keys():
-                foundation_scores[foundation].append(probabilities[0][1].item())
-
+            for idx, label in enumerate(self.label_names):
+                foundation = label.lower()
+                if foundation in foundation_scores:
+                    foundation_scores[foundation].append(probabilities[0][idx].item())
+        
+        # Average the scores across segments
         aggregated_scores = {
             foundation: np.mean(scores) for foundation, scores in foundation_scores.items()
         }
-
+        
         return aggregated_scores
 
     def analyze_emotional_trajectory(self, text, window_size=5):